The paper presents our approach for human-machine interaction with an anthropomorphic mechanical percussionist that can listen to live players, analyze perceptual musical aspects in real-time, and use the product of this analysis to play along in a collaborative manner. Our robot, named Haile, is designed to combine the benefits of computational power, perceptual modeling, and algorithmic music with the richness, visual interactivity, and expression of acoustic playing. We believe that when interacting with live players, Haile can facilitate a musical experience that is not possible by any other means, inspiring users to collaborate with it in novel and expressive manners. Haile can, therefore, serve a test-bed for novel forms of musical human-machine interaction, bringing perceptual aspects of computer music into the physical world both visually and acoustically. Author Keywords Human-robot interaction, music, collaborative systems. ACM Classification Keywords H.5.2 User Interfaces: ...